Emil Velikov [Sat, 15 Aug 2015 14:12:36 +0000 (15:12 +0100)]
vbltest: fix variable shadowing warning
Just remove the second (shadowing) declaration of ret.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Rob Clark [Fri, 4 Sep 2015 15:08:04 +0000 (11:08 -0400)]
freedreno: add API to get drm fd from fd_device
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Fri, 4 Sep 2015 12:08:02 +0000 (08:08 -0400)]
drm: make individual drm_server_info fxns optional
For android / drm_gralloc, we want to hook up our own debug_print()
without bothering with the reset of it.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Thu, 3 Sep 2015 18:56:02 +0000 (14:56 -0400)]
freedreno: don't take ownership of the dmabuf fd on import
EGL_EXT_image_dma_buf_import specifies that the importer retains
ownership of the fd, rather then the importee.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Jonathan Gray [Tue, 1 Sep 2015 11:37:19 +0000 (21:37 +1000)]
amdgpu: use EINVAL instead of EBADMSG in amdgpu_bo_cpu_unmap()
EBADMSG is a streams errno. OpenBSD does not implement streams and does
include the streams errnos, this commit fixes the build on OpenBSD.
None of the callers of this function check the return value for -EBADMSG.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
monk.liu [Mon, 31 Aug 2015 07:06:58 +0000 (15:06 +0800)]
amdgpu: fix missing deinit on vamgr_32
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: monk.liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Varad Gautam [Sun, 30 Aug 2015 10:00:58 +0000 (15:30 +0530)]
freedreno: serialize drmPrimeFDToHandle under table_lock
fixes the prime sharing race condition described by
"intel: Serialize drmPrimeFDToHandle with struct_mutex".
we inline fd_bo_from_handle() into fd_bo_from_dmabuf() and allow locking.
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Varad Gautam [Sun, 30 Aug 2015 10:00:57 +0000 (15:30 +0530)]
freedreno: use drmPrime wrappers to import/export prime bo's
don't call drmIoctl() directly for prime bo's, use the wrappers instead.
v3: remove struct drm_prime_handle and split locking
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Thierry Reding [Wed, 26 Aug 2015 08:05:48 +0000 (10:05 +0200)]
omap: Initialize DMA BUF file descriptor to -1
Commit
c86dabfc9f04 ("omap: zero is a valid fd number, treat it as
such") corrected checks for valid file descriptors, but the OMAP buffer
object code initializes the DMA-BUF file descriptor to 0 (as a result of
calloc()'ing the structure). Obviously this isn't going to work because
subsequent code will try to use file descriptor 0 (most likely stdin at
that point) as a DMA-BUF. It may also try and close stdin when a buffer
object is destroyed.
Fix this by initializing the DMA-BUF file descriptor to -1, properly
marking it as an invalid file descriptor.
Fixes:
c86dabfc9f04 ("omap: zero is a valid fd number, treat it as such")
Reported-by: Robert Nelson <robertcnelson@gmail.com>
Tested-by: Robert Nelson <robertcnelson@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Christian König [Mon, 24 Aug 2015 09:43:30 +0000 (11:43 +0200)]
amdgpu: serialize drmPrimeFDToHandle
Fixes the same problem as "intel: Serialize drmPrimeFDToHandle with struct_mutex".
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Christian König [Mon, 24 Aug 2015 09:43:29 +0000 (11:43 +0200)]
amdgpu: remove sequence mutex
It's not used any more.
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Emil Velikov [Mon, 24 Aug 2015 17:21:51 +0000 (18:21 +0100)]
Move -lm link flag to LIBADD
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Jérôme Glisse [Mon, 24 Aug 2015 17:16:56 +0000 (13:16 -0400)]
drm: Add -lm to libdrm ldflags to fix build failure.
Last commit (
b556ea127e004b734b2a7bf8e67cdcf56312171d) introduced
use of log2 which require -lm flag for the linker on quite few
distribution. Just add that flag to fix build.
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Emil Velikov [Mon, 17 Aug 2015 03:09:06 +0000 (11:09 +0800)]
drm: add interface to get drm devices on the system v3
For mutiple GPU support, the devices on the system should be enumerated
to get necessary information about each device, and the drmGetDevices
interface is added for this. Currently only PCI devices are supported for
the enumeration.
Typical usage:
int count;
drmDevicePtr *foo;
count = drmGetDevices(NULL, 0);
foo = calloc(count, sizeof(drmDevicePtr));
count = drmGetDevices(foo, count);
/* find proper device, open correct device node, etc */
drmFreeDevices(foo, count);
free(foo);
v2: [Jammy Zhou]
- return a list of devices, rather than nodes
v3: [Jammy Zhou]
- fix the signed extension for PCI device info
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Mathias Tillman [Mon, 24 Aug 2015 03:56:13 +0000 (11:56 +0800)]
drm: fix the usage after free
For readdir_r(), the next directory entry is returned in caller-allocted
buffer (pointered by pent here).
https://bugs.freedesktop.org/show_bug.cgi?id=91704
Signed-off-by: Mathias Tillman <master.homer@gmail.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Varad Gautam [Fri, 21 Aug 2015 16:44:35 +0000 (22:14 +0530)]
freedreno: get bo size for imported dma-buf
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Varad Gautam [Fri, 21 Aug 2015 16:44:34 +0000 (22:14 +0530)]
freedreno: fill bo->fd when importing
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Varad Gautam [Fri, 21 Aug 2015 16:44:33 +0000 (22:14 +0530)]
freedreno: fix a bo cache segfault with imported bo's
Importing a bo whose handle is still in the bo cache crashes during cleanup.
Remove bo from cache when importing.
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rafał Sapała [Fri, 24 Jul 2015 09:22:34 +0000 (11:22 +0200)]
intel: Serialize drmPrimeFDToHandle with struct_mutex
It is possible to hit a race condition in create_from_prime, when trying
to import a BO that's currently being freed. In case of prime sharing
we'll succesfully get a handle, but fail on get_tiling call, potentially
confusing the caller (and requiring different locking scheme than with
sharing using flink). Wrap fd_to_handle with struct_mutex to force
a more consistent behaviour between prime/flink, convert fprintf to DBG
when handling errors.
(From Chris:
The race is that the kernel returns us the same file-private handle as
the first thread, but that first thread is about to call gem_close
(thereby removing the handle from the file completely) and does so
between us acquiring the handle and taking the mutex. If we take
the mutex, then we acquire the refcnt on the bo prior to the first
thread completing its unref (and so preventing the early close). Or we
acquire the handle after the earlier close, in which case we are the new
owner.
)
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Testcase: igt/drm_import_export/import-close-race-prime
Signed-off-by: Rafał Sapała <rafal.a.sapala@intel.com>
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Hyungwon Hwang [Wed, 19 Aug 2015 00:58:42 +0000 (09:58 +0900)]
modetest: remove the trailing white spaces
This patch removes the trailing white spaces.
Signed-off-by: Hyungwon Hwang <human.hwang@samsung.com>
Hyungwon Hwang [Wed, 19 Aug 2015 00:58:39 +0000 (09:58 +0900)]
xf86drmMode: remove the trailing white spaces
This patch removes the trailing white spaces.
Signed-off-by: Hyungwon Hwang <human.hwang@samsung.com>
Mauro Rossi [Tue, 18 Aug 2015 09:43:57 +0000 (11:43 +0200)]
amdgpu: add Android build support
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Mauro Rossi [Thu, 20 Aug 2015 15:55:20 +0000 (16:55 +0100)]
amdgpu: add Makefile.sources
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
[Emil Velikov: remove duplicate amdgpu.h from LIBDRM_AMDGPU_FILES]
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Rob Clark [Tue, 18 Aug 2015 15:56:50 +0000 (11:56 -0400)]
Bump version for release
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Tue, 18 Aug 2015 14:53:36 +0000 (10:53 -0400)]
freedreno: update freedreno-symbol-check
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Thierry Reding [Wed, 9 Apr 2014 07:00:49 +0000 (09:00 +0200)]
libdrm: Make indentation consistent
Use tabs and spaces consistently to align function arguments on
subsequent lines with those of the first line.
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Thierry Reding [Wed, 9 Apr 2014 06:59:04 +0000 (08:59 +0200)]
libdrm: Remove gratuitous blank lines
Usage of blank lines can be a matter of taste, of course, but for these
we can surely all agree that they're not needed and inconsistent.
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Jammy Zhou [Mon, 17 Aug 2015 03:09:09 +0000 (11:09 +0800)]
amdgpu: make vamgr per device v2
Each device can have its own vamgr, so make it per device now.
This can fix the failure with multiple GPUs used in one single
process.
v2: rebase
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Mon, 17 Aug 2015 03:09:08 +0000 (11:09 +0800)]
amdgpu: add flag to support 32bit VA address v4
The AMDGPU_VA_RANGE_32_BIT flag is added to request VA range in the
32bit address space for amdgpu_va_range_alloc.
The 32bit address space is reserved at initialization time, and managed
with a separate VAMGR as part of the global VAMGR. And if no enough VA
space available in range above 4GB, this reserved range can be used as
fallback.
v2: add comment for AMDGPU_VA_RANGE_32_BIT, and add vamgr to va_range
v3: rebase to Emil's drm_private series
v4: fix one warning
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Mon, 17 Aug 2015 03:09:07 +0000 (11:09 +0800)]
amdgpu: improve amdgpu_vamgr_init
Make it a generic function independent of the device info.
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Rob Clark [Mon, 17 Aug 2015 14:33:59 +0000 (10:33 -0400)]
freedreno: add fd_pipe_wait_timeout()
We need to pass through a timeout parameter to implement
pipe->fence_finish() properly.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Tue, 21 Jul 2015 16:55:29 +0000 (12:55 -0400)]
freedreno/msm: dump out submit info on error
User should only see these with LIBGL_DEBUG=verbose. But in case you
are hitting issues like "handle X at index Y already on submit list"
errors from the kernel, this gives some useful visibility for debug.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Tue, 21 Jul 2015 15:57:00 +0000 (11:57 -0400)]
freedreno/msm: fix issue where same bo is on multiple rings
It should be a less common case, but it is possible for a single bo to
be on multiple rings, for example when sharing a buffer across multiple
pipe_context's created from same pipe_screen.
So rather than completely fall over in this case, fallback to slow-path
of looping over all bo's in the ring's bo-table (but retain the fast-
path of constant-lookup for the first ring the buffer is on).
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Tue, 21 Jul 2015 16:11:36 +0000 (12:11 -0400)]
freedreno/msm: reorg ringbuffer struct
Group the parts related to building out submit ioctl into their own
sub-struct. Split out from next commit since it is just boring churn.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Michel Dänzer [Mon, 17 Aug 2015 09:43:39 +0000 (18:43 +0900)]
tests/amdgpu: Remove unused local variable 'i'
Reviewed-by: Christian König <christian.koenig@amd.com>
Michel Dänzer [Mon, 17 Aug 2015 09:41:11 +0000 (18:41 +0900)]
tests/amdgpu: Include config.h first
Fixes build failure on 32-bit because _FILE_OFFSET_BITS wasn't defined to
64.
Reviewed-by: Christian König <christian.koenig@amd.com>
Thierry Reding [Fri, 23 Jan 2015 16:08:21 +0000 (17:08 +0100)]
tests: modetest: Accept connector names in addition to connector IDs
Allow connector names to be used in the specification of the -s option.
This requires storing the string passed on the command-line so that it
can later be resolved to a connector ID (after the DRM device has been
opened).
Connector names are constructed from the connector type name and
connector type ID using the same format as used internally in the
Linux kernel.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Marek Olšák [Fri, 14 Aug 2015 12:19:29 +0000 (14:19 +0200)]
Bump version for release
Emil Velikov [Fri, 7 Aug 2015 16:29:11 +0000 (17:29 +0100)]
amdgpu: add symbols check test
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 16:09:35 +0000 (17:09 +0100)]
amdgpu: hide the final internal functions from global namespace
Thus the only symbols that we export are the ones officially provided by
the API.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 15:54:29 +0000 (16:54 +0100)]
amdgpu/amdgpu_vamgr: hide private symbols from global namespace
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 15:48:02 +0000 (16:48 +0100)]
amdgpu: squash trivial documentation typo
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 16:20:51 +0000 (17:20 +0100)]
amdgpu: cosmetic chances in license boilerplate
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 16:17:43 +0000 (17:17 +0100)]
amdgpu: add a bunch of missing config.h includes
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 15:44:33 +0000 (16:44 +0100)]
amdgpu/util_hash_table: hide private symbols from global namespace
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 15:40:45 +0000 (16:40 +0100)]
amdgpu/util_hash: hide private symbols from global namespace
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Emil Velikov [Fri, 7 Aug 2015 15:19:49 +0000 (16:19 +0100)]
Force enable amdgpu for the dist build/check.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Alex Deucher [Mon, 10 Aug 2015 19:37:19 +0000 (15:37 -0400)]
radeon: add new OLAND pci id
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tapani Pälli [Fri, 7 Aug 2015 07:37:58 +0000 (10:37 +0300)]
modeprint: cleanup, remove compile warnings
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Tapani Pälli [Fri, 7 Aug 2015 07:37:57 +0000 (10:37 +0300)]
intel: wrap intel_bufmgr.h C code for C++ compilation/linking
We need this include in porting changes for the OpenGL ES
conformance suite.
v2: remove c_plusplus usage
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Tapani Pälli [Fri, 7 Aug 2015 07:37:56 +0000 (10:37 +0300)]
remove usage of 'c_plusplus' preprocessor macro
Use only __cplusplus which is supported by the C++ standard.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Alex Deucher [Thu, 6 Aug 2015 17:44:16 +0000 (13:44 -0400)]
add a note about which version of cunit is fixed for debian/ubuntu
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 6 Aug 2015 17:38:31 +0000 (13:38 -0400)]
move up cunit workaround for ubuntu/debian
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 6 Aug 2015 15:19:55 +0000 (11:19 -0400)]
fix amdgpu cunit configure test harder
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 6 Aug 2015 15:13:48 +0000 (11:13 -0400)]
fix configuration when amdgpu is disabled
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Tue, 4 Aug 2015 03:37:23 +0000 (12:37 +0900)]
amdgpu: Remove unused local variables bo_size/offset from amdgpu_cs_submit
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Jammy Zhou [Mon, 3 Aug 2015 07:46:42 +0000 (15:46 +0800)]
amdgpu: expose the PCI revision ID
The PCI revision ID can be used to differentiate ASICs.
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Mario Kleiner [Wed, 29 Jul 2015 19:09:04 +0000 (21:09 +0200)]
libdrm/amdgpu: Use private fd for amdgpu_device and winsys hash table to fix ZaphodHeads. (v2)
The amdgpu_device for a device node needs its own dup'ed fd, instead
of using the original fd passed in for a screen, to make multi-x-screen
ZaphodHeads configurations work on amdgpu.
The original fd's lifetime differs from that of the amdgpu_device, and from the
one stored in the hash. The hash key is the fd, and in order to compare hash
entries we fstat them, so the fd must be around for as long as the amdgpu_device
is.
This patch for libdrm/amdgpu is a translation of the radeon-winsys ZaphodHeads
fix for mesa's radeon-winsys, from mesa commit
28dda47ae4d974e3e032d60e8e0965c8c068c6d8
"winsys/radeon: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads."
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Acked-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
v2: Check for valid fd's being >= 0, because fd == 0 is in theory
a valid, although unlikely, fd and fd == -1 would denote an
invalid fd. Thanks to William Lewis for pointing this out.
Reported-by: William Lewis <minutemaidpark@hotmail.com>
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Alexandr Akulich [Thu, 16 Jul 2015 20:19:12 +0000 (01:19 +0500)]
libdrm/amdgpu: Fixed drm.h include.
The include type changed from system to own.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alexandr Akulich <akulichalexander@gmail.com>
Michel Dänzer [Tue, 21 Jul 2015 08:53:42 +0000 (17:53 +0900)]
amdgpu/tests: Cast CPU map argument of amdgpu_bo_alloc_and_map to void**
Silences incompatible pointer type compiler warnings.
Reviewed-by: Christian König <christian.koenig@amd.com>
Michel Dänzer [Tue, 21 Jul 2015 08:51:27 +0000 (17:51 +0900)]
amdgpu/tests: Use buf_handle in amdgpu_bo_alloc_and_map() error paths
The compiler pointed out that bo has the wrong type.
Reviewed-by: Christian König <christian.koenig@amd.com>
Leo Liu [Mon, 13 Jul 2015 16:51:34 +0000 (12:51 -0400)]
amdgpu: add VCE harvesting instance query
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Ken Wang [Fri, 17 Jul 2015 11:39:56 +0000 (19:39 +0800)]
test/amdgpu : fix a bug in VCE/UVD test introduced by previous change
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Jammy Zhou [Mon, 13 Jul 2015 12:57:44 +0000 (20:57 +0800)]
amdgpu: add amdgpu_bo_va_op for va map/unmap support v3
The following interfaces are changed accordingly:
- amdgpu_bo_alloc
- amdgpu_create_bo_from_user_mem
v2: update the interfaces
v3: remove virtual_mc_base_address from amdgpu_bo
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Thu, 16 Jul 2015 02:29:58 +0000 (10:29 +0800)]
amdgpu: add flags parameter for amdgpu_va_range_alloc
The flags is added for extensibility to cover some special requirements
in the future, i.e, request VA range in the first 4GB of address space
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Ken Wang [Fri, 10 Jul 2015 14:22:27 +0000 (22:22 +0800)]
amdgpu : move management of user fence from libdrm to UMD
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Wed, 15 Jul 2015 08:56:20 +0000 (10:56 +0200)]
amdgpu: fix bs buffer size for vce test
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Jammy Zhou [Mon, 13 Jul 2015 14:43:29 +0000 (22:43 +0800)]
tests/amdgpu: remove the duplicate IB allocation for VCE test
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Mon, 13 Jul 2015 08:49:42 +0000 (16:49 +0800)]
drm: fix the ALIGN macro to avoid value clamp
If the value is 64bit, but the alignment is 32bit type, the high 32bit
will be clamped with previous definition
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Michel Dänzer [Tue, 14 Jul 2015 03:42:19 +0000 (12:42 +0900)]
amdgpu: Use drmIoctl in amdgpu_ioctl_wait_cs
This is safe now because the ioctl uses an absolute timeout.
This prevents amdgpu_cs_query_fence_status from returning early e.g.
when a signal is delivered, which in turn caused Mesa winsys code to
assume a BO was idle when it actually wasn't yet.
Reviewed-by: Christian König <christian.koenig@amd.com>
Christian König [Thu, 9 Jul 2015 09:48:32 +0000 (11:48 +0200)]
amdgpu: use common fence structure for dependencies as well.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Jammy Zhou [Thu, 9 Jul 2015 05:51:13 +0000 (13:51 +0800)]
amdgpu: improve the amdgpu_cs_query_fence_status interface
make amdgpu_cs_query_fence reusable to support multi-fence query
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Sabre Shao [Thu, 9 Jul 2015 05:53:24 +0000 (13:53 +0800)]
amdgpu: add va range query interface
amdgpu_va_range_query interface is added so that client can query va
range supported by specific device.
Signed-off-by: Sabre Shao <Sabre.Shao@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Sabre Shao [Thu, 9 Jul 2015 05:50:36 +0000 (13:50 +0800)]
amdgpu: add va allocation intefaces
Two new interfaces are added to support client request for
allocate virtual address without physical memory committed to.
The virtual address space can be managed by client itself.
Signed-off-by: Sabre Shao <Sabre.Shao@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Ken Wang [Thu, 9 Jul 2015 05:48:25 +0000 (13:48 +0800)]
amdgpu: add base_preferred parameter to amdgpu_vamgr_find_va
base_preferred parameter is added to amdgpu_vamgr_find_va
so UMD can specify preferred va address when allocating.
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Marek Olšák [Wed, 8 Jul 2015 09:12:31 +0000 (11:12 +0200)]
amdgpu: check the user fence only if the IP supports user fences
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Marek Olšák [Fri, 26 Jun 2015 19:58:17 +0000 (21:58 +0200)]
drm/amdgpu: allow passing absolute timeouts to amdgpu_cs_query_fence_status
Useful when Mesa wants to wait for a lot of fences at the same time and
doesn't want to recalculate the relative timeout after every call.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Tue, 30 Jun 2015 14:27:27 +0000 (16:27 +0200)]
amdgpu: remove reference to AMD specific error codes
We just have never defined any.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Tue, 30 Jun 2015 14:04:44 +0000 (16:04 +0200)]
amdgpu: cleanup public interface style
Fix some style problems, adjust to a common indentation, reorder two
function definitions and remove stale comments.
No intended functional change.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Tue, 30 Jun 2015 13:46:29 +0000 (15:46 +0200)]
gitignore: add some generated amdgpu files
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Wed, 24 Jun 2015 12:17:57 +0000 (14:17 +0200)]
amdgpu: add CS dependencies v2
This allows the driver to specify on which previous CS to wait.
v2: fix spelling in comment
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Alex Deucher [Mon, 29 Jun 2015 16:02:47 +0000 (12:02 -0400)]
amdgpu: update to the latest kernel header
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 24 Jun 2015 12:13:00 +0000 (14:13 +0200)]
amdgpu: remove pointer arithmetic from command submission
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Jammy Zhou [Tue, 16 Jun 2015 14:42:43 +0000 (22:42 +0800)]
amdgpu: do NULL check for bo handle in amdgpu_bo_query_info
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Christian König [Mon, 8 Jun 2015 13:05:07 +0000 (15:05 +0200)]
amdgpu: cleanup VA IOCTL handling
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Sat, 6 Jun 2015 10:02:29 +0000 (12:02 +0200)]
amdgpu: remove flink export workaround v2
Alternative solution to Mareks patch to stop causing trouble with render nodes.
v2: rebased
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Sat, 6 Jun 2015 10:00:17 +0000 (12:00 +0200)]
amdgpu: explicitly unmap GPU mapping on BO destruction
That exercises the IOCTL and stops relying us on implicit unmapping.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 5 Jun 2015 21:07:56 +0000 (05:07 +0800)]
amdgpu: merge amdgpu_drm.h from kernel
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 5 Jun 2015 21:02:23 +0000 (05:02 +0800)]
amdgpu: cleanup gds specific alloc/free functions
amdgpu_bo_alloc/free can be used with GDS/OA/QWS domains specified
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 5 Jun 2015 21:00:36 +0000 (05:00 +0800)]
amdgpu: add amdgpu_query_gds_info
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Thu, 4 Jun 2015 16:57:57 +0000 (18:57 +0200)]
amdgpu: fix double mutex_unlock in amdgpu_bo_import
The handles array is used below, which is followed by another unlock,
so remove the first one.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Thu, 4 Jun 2015 16:57:50 +0000 (18:57 +0200)]
amdgpu: fix valgrind warnings
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Tue, 2 Jun 2015 12:43:52 +0000 (14:43 +0200)]
amdgpu: use alloca and malloc in critical codepaths (v2)
And don't clear the memory when it's unnecessary.
v2: use malloc for arrays that can be big
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Thu, 4 Jun 2015 10:47:22 +0000 (12:47 +0200)]
amdgpu: allow exporting KMS handles with render nodes
Reviewed-by: Christian König <christian.koenig@amd.com>
Jack Xiao [Thu, 4 Jun 2015 02:41:24 +0000 (10:41 +0800)]
amdgpu: add zero timeout check in amdgpu_cs_query_fence_status
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Marek Olšák [Tue, 2 Jun 2015 11:05:41 +0000 (13:05 +0200)]
amdgpu: remove bo_handle from amdgpu_cs_ib_info, IBs should be in buffer list
Reviewed-by: Christian König <christian.koenig@amd.com>
Marek Olšák [Mon, 1 Jun 2015 12:12:10 +0000 (14:12 +0200)]
amdgpu: remove amdgpu_ib helpers
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Marek Olšák [Fri, 29 May 2015 17:13:41 +0000 (19:13 +0200)]
amdgpu: remove amdgpu_ib
Not useful if we're gonna use BO handles directly.
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Marek Olšák [Fri, 29 May 2015 15:13:12 +0000 (17:13 +0200)]
amdgpu: don't use amdgpu_cs_create_ib for allocation of the fence BO
amdgpu_cs_create_ib will go away.
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Mon, 25 May 2015 12:38:03 +0000 (20:38 +0800)]
tests/amdgpu: manage IB in client side
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>